The hallmark of spatial memory is the ability to link objects or landmarks with their spatial locations. Significant failures of spatial memory are among the first major clinical markers of neurodegenerative disease. The proposed research aims to illuminate the neural and cognitive mechanisms underlying human spatial memory through the analysis of electrocorticographic and single-neuron recordings taken as neurosurgical patients navigate through virtual environments in search of target objects or locations. The proposed studies will examine the relation between brain waves (oscillations), cellular responses, and subjects' behavior during several types of game-like virtual navigation tasks. Our first two aims focus on the coding of spatial information. Aim 1 seeks to answer the question: How is the information necessary for (virtual) spatial navigation and spatial memory represented by neuronal activity in various brain regions? Aim 2 attempts to complement this knowledge by asking the question: How do these spatially-relevant neural systems code information? The third and fourth aims regard the spatial memory system as a dynamic entity, and investigate the ways in which the responses of this system change over time, due to learning and to modifications of the environment. Aim 3 addresses the question of how spatial representations are acquired and transformed through experience. Aim 4 investigates the interaction between spatial memory and verbal episodic memory. The research supporting this aim will enable us to link our research on spatial memory to the larger literature on the role of the medial-temporal-lobe system in declarative memory processes in humans and animals. The proposed studies are of direct relevance to the treatment of epilepsy, in which mapping of cognitive functions to brain regions during surgical procedures is crucial for ensuring good postsurgical outcome, as well as developing treatments for other disorders of memory and cognition.